A study of artificial neural networks for estimating riverine biodiversity

As the population and demand for land use rapidly increased, the use of environmental resources has exceeded the rate of naturalization that might result in the degeneracy of ecological structure and the decrease of the diversities of species which could reduce the resources provided by environment. Due to the raise of eco-environmental restoration concept in the past years, people gradually pay attention to the coexistence relationship between human beings and eco-environment and the impacts of human activities on eco-environment. Stream flow management is the idea that combines the concept of ecology and provides the demand for both human and river ecosystem. This study built the Self-Organizing Radial Basis Neural Networks to categorize the stream flow data and estimate the diversities of fish families in river ecosystem by using Taiwan Ecohydrology Index System (TEIS). In this study there are 60 flow stations, and the stream flow data were only collected with records more than 20 years. A moving average method was applied to TEIS statistics to reflect the effects of antecedent flow condition. The input data are the TEIS statistics of different moving average periods. The output of the model showed a high authenticity of the prediction for the diversities. The result shows that this model can not only categorize the stream flow data but also estimate the bio-diversity efficiently and precisely.